A Photoelectron-Photoion Coincidence Study of ... - ACS Publications

Rev., 151, 207 (1966). A Photoelectron-Photoion Coincidence Study of the Ionization and Fragment. Appearance Potentials of Bromo- and lodomethanes...
0 downloads 0 Views 576KB Size
570

E. P. Tsai, T. Baer, A. S. Werner, and S. F. Lin

the mixture, the formation of the cyclohexyl radical is suppressed, while t-C4H9 radical is produced complementary by the addition of H atoms to i ~ 0 b u t e n e . lThe ~ formation of the neopentyl radical is not affected by the addition of isobutene. (4) Temperature and isotope effects are also observed in neopentane-alkane mixtures (Tables I and 11). These effects are quite similar to those inkhe hydrogen abstraction reaction of H atoms. Therefore the mechanism of the formation of solute alkyl radical in the radiolysis of neo-C~Hl2-i-C4HgDa t 77 K may be represented as follows: neorC5H12vv+ C,Hl,* (7)

-

C5H1,* --+ neo-C,H1,

H

112)

I-C,H,D t-C,H, + HD (4) H atoms produced by the fragmentation of excited neopentane molecules react selectively with i-CdH9D to form tC4H9 radicals and HD. It was reported that the radiolysis of solid 3-methylpentane-dl4 produces no trapped D atoms,ls while trapped H atoms are observed in the radiolysis of solid CH4.7J9 The formation of H atoms in the radiolysis of solid neopentane does not coincide with the result in the radiolysis of 3-methylpentane glass. It is not clear a t present that the reactive H atoms in reaction 12 are the same ones as the trapped H atoms. This problem requires further studies in future. According to the reaction scheme, the yield of the t-C,Hg radical should be equal to that of the neo-CaH11 radical. Table I1 shows, however, that the yield of the solute radical is much higher than that of the neo-CsH11 radical. Therefore we must consider another origin of hydrogen atoms, for example C,HI2* + C,H,, C,,H,, + 2H (13) H

+

-1-

-

Acknowledgment. The authors wish to express their appreciation to Professor Zen-ichiro Kuri of Nagoya University for his encouragement, and to Dr. Kozo Matsumoto of Nagoya University for mass spectrometric analysis.

References and Notes (1) (a) T. Miyazaki, T. Wakayama, K. Fueki, and Z. Kuri, Bull. Chem. SOC. Jap., 42, 2086 (1969); (b) T. Wakayama, T. Miyazaki, K. Fueki, and 2. Kuri, J. Phys. Chem., 74, 3584 (1970); (c) Y. Saitake, T. Wakayama, T. Kimura, T. Miyazaki, K. Fueki, and 2. Kuri, Bull. Chem. SOC. Jap.. 44, 301 11971). (2) T. Wakayama, T. Miyazaki, K. Fueki, and 2.Kuri, Bull. Chem. SOC.Jap., 44. 2619 11971). (3) M.'Fukaya, T. Wakayama, T. Miyazaki, Y. Saitake, and Z. Kuri, Bull. Chem. SOC.Jap., 46, 1036 (1973). (4) (a) T. Miyazaki, T. Wakayama, M. Fukaya, Y. Saitake, and 2. Kuri, Bull. Chem. SOC.Jap., 46, 1030 (1973); (b) M. Kato, Y. Saitake, T. Miyazaki, and Z. Kuri, ibid., 46, 2004 (1973). (5) D. J. Henderson and J. E. Willard, J. Amer. Chem. SOC., 91, 3014 (1969). (6) T. Wakayama. T. Miyazaki. K. Fueki. and 2. Kuri, J. Phys. Chem., 77, 2365 (1973). (7) 8. Smaller and M. S. Matheson, J. Chem. Phys., 28, 1169 (1958). (8) J. Roncin, Mol. Cryst., 3, 117 (1967). (9) F. Thyrion, J. Dodelet, C. Fauquenoit, and P. Claes, J. Chim. Phys., 65, 227 (1968). (IO) W. G. French and J. E. Willard, J. Phys. Chem., 72, 4604 (1968). (11) L. Perkey and J. E. Willard, J. Chem. Phys., 60, 2732 (1974). (12) J. H. Sullivan, J. Chem. Phys., 30, 1292 (1959); 36, 1925 (1962). (13) B. A. Thrush, Progr. React. Kinet., 3, 89 (1965). (14) (a) E. D. Sprague and F. Williams, J. Amer. Chem. SOC., 93, 787 (1971); (b) R. J. LeRoy, E. D. Sprague, and F. Williams, J. Phys. Chem., 76, 546 (1972); (c) A. Campion and F. Williams, J. Amer. Chem. SOC., 94, 7633 (1972); (d) E. D.Sprague, J. Phys. Chem., 77, 2066 (1973). (15) J. Lin and F. Williams, J. Phys. Chem., 72, 3707 (1968). (16) W. H. Taylor, S. Mori, and M. Burton, J. Amer. Chem. SOC., 82, 5817 (1960). (17) M. Kato, T. Miyazaki, and 2. Kuri, unpublished results. (18) D. Timm and J. E. Willard, J. Phys. Chem., 73, 2403 (1969). (19) (a) D.W. Brown, R. E. Florin, and L. A. Wahl, J. Phys. Chem., 86, 2602 (1962); (b) W. Gordy and R. Morehouse, Phys. Rev., 151, 207 (1966).

A Photoelectron-Photoion Coincidence Study of the Ionization and Fragment Appearance Potentials of Bromo- and lodomethanes Bilin P. Tsai, Tomas Baer,' Arthur S. Werner, and Stephen F. Lin William R. Kenan, Jr. Laboratory of Chemistry, University of North Carolina, Chapel Hill, North Carolina 27514 (Received September 23, 1974)

The ionization and fragment appearance potentials of CHsBr, CH2Br2, CHBr3, CH31, CH212, and CHI3 have been derived from an analysis of zero kinetic energy photoelectron-photoion coincidence spectra in the energy range 9-14 eV. Such coincidence spectra yield more precise appearance potentials than integral photoionization scans because of the inherently sharper onsets and the discrimination against competing processes such as ion-pair formation and collisionally induced dissociation.

Introduction The ionization and fragmentation of the halogenated alkanes exhibit a number of unusual properties. The highest filled molecular orbitals have the predominant character of the halogen lone pairs and the resulting first photoelectron bands exhibit the features which are characteristic of the The Journal of Physical Chemistry, Vol. 79, No. 6, 7975

removal of nonbonding electrons.lT2 Nevertheless the molecular ions, which under these circumstances might be expected to be very stable, in fact readily fragment. Both the C-X and the C-H bond energies decrease with increasing halogen substitution resulting, in the case of carbon tetrachloride3 and carbon tetrafluoride: in unstable parent ions. The driving force for this trend is the stability of the

Ionization and Appearance Potentials of Bromo- and lodomethanes planar halogenated methyl ions which have a closed shell electronic structure. A characteristic feature in the fragmentation of the chloromethane ions produced by photoionization below 13.5 eV is the absence of hydrogen loss even though this reaction in CH3Ci is energetically favored over chlorine loss. In all the other chloro-, bromo-, and iodomethanes, halogen atom loss is energetically favored, and in two molecules (CH3Br and CHJ) which we studied with sufficient mass resolution to resolve the H loss fragment, we again found no evidence for H loss. w e can rule out an activation barrier because electron impact experiments such as those of Losing5 produce strong onsets for hydrogen loss in CH3C1+ at precisely the thermodynamic appearance potential. This point is discussed further elsewhere.6 Little work has been done on the ionization and especially the fragment appearance potentials of the bromo- and iodomethanes. There are two investigations in which the photoelectron spectra (PES) of all the molecules in this study were obtained.132 In addition other workers have reported PES of the mon~halomethanes.~-~ Several electron impact investigations have been reported,lo-ls however, none have dealt with CH& or CHI3 and only one16 with CHBr3. With the exception of one study,lg the available photoionization spectra (PI) have been measured without mass analysis and at low photon resolution.20-22 As a result much of the published work concerns only the parent ionization potentials. We have undertaken this study of the bromo- and iodomethanes in order to establish reliable values for fragment ion appearance potentials and heats of formation.

571 E+E-Eo

PY(E) = l m d c

dx g ( c ) e - ' l k T p ( x ) ( 1 ) 0

0

ground state molecule, and p ( x ) is the ionization transition probability at an energy, x , above the ionization potential. Eo is the adiabatic ionization or fragment appearance potential. The density of states can be determined by direct count, and p ( x ) is known from PES. By using the PES for p ( r ) we neglect the contribution of autoionization, an assumption which can be justified only for certain systems. For example, the PI spectrum of CH3IZ4shows numerous autoionization peaks between threshold and 0.7 eV above the first ionization potential. However, for the molecules studied here, the fragmentation onsets were in regions which exhibited relatively little autoionization. In the present study we found that the hot bands contribute significantly more to the total ionization than they did in the chloromethanes. In addition, autoionization increases with increasing mass of the halogen. As a result the analysis of the bromo- and iodomethanes was considerably more difficult. In order to sharpen up the onsets as well as to discriminate against autoionization and possible competing processes such as ion-pair formation and collisionally induced fragmentation we analyzed the ZKE electronion coincidence curves. For such curves, eq 1 reduces to eq 2 in which integration over the distribution of ejected electron energies is eliminated.

CY@)

=

Jm

0

de g(c)e-"kTp(E

+ E - E,)

(2)

Experimental Section The apparatus, presented in detail in a previous paper,3 will only briefly be described here. A hydrogen "many-line" light source dispersed by a 1-m vacuum uv monochromator to give 2-8, resolution ionized the molecules in a photoionization chamber; the sample gas pressure was approximately loh4 Torr. Ions were mass analyzed with a quadrupole mass filter and/or by time of flight. A nominally zero kinetic energy (ZKE) electron analyzerz3 was used to provide both the "start" of the ion time of flight as well as to obtain ZKE photoelectron spectra. The ion flight time distribution was displayed on a multichannel pulse height analyzer and the appropriate region was read out. During a scan, ZKE electrons, total ions, ZKE electron-ion coincidences, as well as photon intensity and sample gas pressure were counted for preset time intervals. The data were reduced by normalizing to the photon intensity and sample gas pressure. All spectra were taken at room temperature. The quadrupole mass filter was used only on CH3Br and CH31 because the masses of the other molecules exceeded the limit of our quadrupole.

Figures 1-4 show the ZKE electron-ion coincidence spectra in which the counts are normalized to the total light intensity. A summary of the derived ionization and fragment appearance potentials of the molecules investigated is shown in Tables I and I1 along with the results of other workers. The values for the heats of formation were derived without regard to a possible correction due to kinetic energy release. These values are thus upper limits. The kinetic energy release for C1 loss from CH2C12+ has been measured by Cooks et al.25 and found to be less than 0.015 eV. As in CH2ClZ+, both CH2Br2+ and CH&+ show collisional dissociation to give CH2X+ as evidenced by the long slowly rising tails on the daughter P I curves below the true appearance potentials. The intensity of this bimolecular contribution decreases with decreasing sample gas pressure, but due to the high cross section for collisional dissociationZ6the tails are evident even in the lowest pressure runs. Although the tail cannot be totally suppressed in the ion curve, the coincidence curves do effectively suppress the contribution of the bimolecular reaction because of the displaced flight time of the collisionally induced CH2X+ fragments.

Results

Discussion

In our previous study3 of the chloromethanes and carbon tetrabromide, we developed a general procedure for fitting calculated photoionization spectra to the experimental curves. The purpose of this was to take into account the contribution of hot bands and variations in the ionization transition probability in calculating ionization and fragment appearance potentials. This procedure is particularly necessary for accurately determining the latter. The calculated photoionization yield, PY, a t a photon energy, E, is given by eq 1 in which g(c) is the density of states of the

CH3Br. The photoionization curve of methyl bromide is well knownlg-21 and will not be presented in this paper. The first ionization potential is due to the ejection of an electron from a nonbonding 4pre orbital localized on the bromine atom. The onset is very sharp and gives an ionization potential, 10.54 eV, in good agreement with those obtained from photoelectron, photoionization, and absorption spectroscopy (Table I). The coincidence spectrum for the CH3+ fragment is similar to the photoionization spectrum of Krausslg except that our onset is sharper. Our threshold The Journal of Physical Chemistry, Vol. 79, No. 6, 1975

B. P. Tsai, T. Baer, A. S. Werner, and S. F. Lin

572

TABLE I: First Ionization Potentials CH,Br' Apf(298), kcal mo1-l"

234.6

i. 0.3

CH,Br,+ 241.5

i

CHBr,+

1.5

245.6

Technique

i 0.4

CH,I+ 222.9

i 0.3

CH&'

CHI,'

245.1

f 0.6

263.2

f 0.4

9.46

0.02b

9.25

i 0.02b

9.26

i 0.02'

Ionization potentials, eV

Photoionization

10.54 i O . O l b 10.528 i 0.005" 10.53 f O.Old*e

10.52 10.49

Photoelectron spectroscopy

10.53 i O.Olh 10.53 i 0.02i 10.53 i 0.03j 10.54k'' 10.5m*" 10.53 i 0.02O 10.56e 10.8 f 0.Ol4 10.54 & 0.04"

10.61k 10.63'

Electron impact Penning ionization Absorption spectroscopy

f 0.05b i 0.02'

10.48 10.51

i 0.026 f 0.02d

10.47k 10.44'

10.8 rt O.Ola

9.533 i O.OlbVf 9.54 f O.Old 9.550 f 0.006' 9.55 i o.lg 9.50 i 0.03j 9.52 i 0.02i 9.54k 9.55' 9 .50m9 9.51 rt 0.02O 9.59* 9.55

f

f

9.46' 9.52'

0.04"

10.488 i 0.003' 10.541 i 0.003t

9.49 i 0.003' 9.538t 9.54' Based on our ionization potentials and the following heats of f o r m a t i ~ n : CH3Br ' ~ ~ ~ ~(-8.4); CHzBrz (-1); CHBr3 (4);Br (26.74);CH3I (3.1); CHzIz (27); CHI3 (50);I (25.5); H (52.1). The values are in kcal mol-'. This work. Reference 21. Reference 20. e Reference 19. f Reference 24. Reference 22. Reference 7. Reference 8. Reference 9. Reference 1. Reference 2. Reference 14. Reference 15. O Reference 11.P Reference 12. q Reference 13. Reference 29. Reference 28. Reference 27. Reference 31. J

TABLE 11: X Loss Appearance Potentials CH,'/CH,Br AH", (2983, kcal mol-' a

260.0

i 0.6

CH,Br+/CH,Br, CHBr,'/CHBr, 234.0

i 0.4

Technique

223.9

0.4

CH,'/CH,I

~ H , I + ~ C H , I , CHI,'/CHI,

260.0

244.6

i 0.6

i 0.4

249.7

i 0.4

Appearance potentials, eV

Photoionization Electron impact

12.80 12.77'

i: 0.03b

11.35 rt 0.02* 10.93 f 0.04d 11.5 i O.Ole

10.70 10.80

12.25 f 0.03b 10.55 f 0.02b 9.77 i 0.02b 12.2 i 0.05f 12.22 f 0.03g 12.36 i 0.02h 12.4 f 0.2i Based on our appearance potentials and the neutral heats of formation listed under Table I. This work. Reference 19. Reference 16. e Reference 13. f Reference 15. g Reference 18. Reference 17. Reference 14. analysis yields an appearance potential of 12.80 eV, which is precisely the thermodynamic onset for Br loss. As mentioned previously, only bromine loss is detected, however, the appearance potential (AP) of CH2Br+ from CHSBr+ can be determined from a knowledge of the heat of formation of CH2Br+ (see Table 11).According to eq 3, the

+ AH', (CH2Br+)+ KE = 12.77 eV + KE released

AP(CH,Br+) = AH", (H) AH",(CH,Br)

(3)

thermodynamic AP of CH2Br+ neglecting kinetic energy release is 12.77 eV, which makes it nearly isoenergetic with Br loss. CH2Br2. Figure 1 shows the coincidence curves for CH2Br2+ and CH2Br+. The fourfold degenerate halogen lone pair electron states are split into three peaks which are poorly resolved in Figure 1. However three peaks are clearly visible in the PES132 which indicates that one is a composite of two states. It was impossible to fit the CHzBrz+ The Journal of Physical Chemistry, Vol. 79, No. 6, 1975

~t0.02b

i

O.Old

coincidence curve with our fitting procedure and as a result no reliable IP was derived. The slow onset between 10.4 and 10.5 eV could possibly be attributed to autoionization from hot bands, although Price's P E S also exhibits an anomalously slow onset. Our best guess is that the IP of CHzBrz is 10.52 f 0.05 eV. The low value is chosen partly to allow the ionization potentials of successively higher substituted methyl bromides to monotonically decrease in a manner similar to that in the series of methyl iodides. The CH2Br+ onset was very sharp and strong. The thermodynamic appearance potential for H loss is 12.02 eV. However lack of sufficient mass resolution did not allow us to determine whether H loss does occur. CHBr3. In the bromoform ion the 4pr, degenerate state splits into four states, two of which remain degenerate. The coincidence curve, shown in Figure 2, was obtained by time-of-flight mass analysis of the ion product in which the photoelectron and photoion provided the start and stop signals, respectively. The coincidence scan shows only the

Ionization and Appearance Potentials of Bromo- and lodomethanes

10,OO

10.50

11,OO 11,50 PHOTON ENERGY (EV)

12,OO

573

12,5

9.5

Flgure 1. The ZKE electron-ion coincidence spectrum of CH2Br2 and its fragment. The parent ion data beyond 11.3 eV is not plotted because it neariy overlaps the daughter ion signal. The points are experimental values and the solid lines are calculated fits. The parent ion curve could not be fit. (See Text.)

Figure 3. The ZKE electron-ion coincidence spectrum of CH2I2 and

its fragment. The points are experimental values and the solid lines are calculated fits.

9,o

10.2

10,4

10,6

l0,B

11,o

11,2

PHOTON ENERGY (EV)

Figure 2. The ZKE electron-ion coincidence spectrum of CHBr3 and 'its fragment. The points are experimental values and the solid lines

are calculated fits.

first state and the calculated fit gives an IP of 10.48 eV, 40 meV above Turner's PES peak.2 Our value is further supported by a reported PES value1 of 10.47 eV. This leads us to suspect the low value which Turner assigned to the first PES peak. The slow rise in the threshold region indicates a substantial hot band contribution. The peak at 10.65 eV is assigned to AI since there is no electronic state observedlJ in that region. The fragment ion threshold, like that of the parent, rises very gradually and the derived appearance potential is 10.70 eV. The thermodynamic appearance potential for H loss from CHBr3+ was calculated to be 12.23 eV using a value3 for AH"f (CBr3+) of 233.7 kcal mol-l however lack of high-resolution mass analysis prevented us from distinguishing the parent ion from the H loss fragment ion. CH3I. The photoionization curve of CH3I has been published by several w o r k e r ~ , l ~ and - ~ most ~ recently by usz4 and will not be shown here. Like CHBCl and CH3Br, the first electron ejected is from the valence npre orbital localized on the iodine atom. Our ionization potential, 9.533 eV, agrees well with the spectroscopic value,27 9.538 eV. The

L O

10.0 10,5 PHOTON ENERGY (EV)

9,2

9,4

9.5

9 s

iO.0

PHOTON E Y E K Y (EL?

Figure 4. The ZKE electron-ion coincidence spectrum of CHI3 and Its fragment. The points are experimental values and the solid lines are calculated fits.

second state of the doublet occurs a t 10.158 eV giving a spin-orbit separation of 0.622 eV, in precise agreement with Price's value, 0.622 eV. There is a great deal of autoionization from Rydberg states of the series converging to the upper state, 2E1/2. Although the I loss produced a strong CH3+ onset a t 12.25 eV, no H loss was observed, the thermodynamic onset for the latter process being 12.74 eV. CHzIz. The coincidence curves of CH&+ and CHzI+ were obtained by TOF mass analysis of the ions and are shown in Figure 3. Besides two PES1,2studies there have been no other data reported for the ionization and fragment appearance potentials of CH&. Like CHZBr2, four states result from the splitting of the degenerate 4pr, state: bz a t 9.5 eV; a2 at 9.8 eV; a1 at 10.3 eV; and bl at 10.6 eV.lJ The first two electronic states are shown in the parent coincidence curve. The best fit to this curve in the threshold region yields an IP of 9.46 eV. There is a great deal of autoionization: the shoulder on the low-energy side of the threshold, as well as the structure from 9.9-10.3 eV, are assigned to autoionization from initially excited vibrational states of the neutral. The second electronic state occurs a t 9.80 eV which agrees well with the PES value. The Journal of Physical Chemistry, Vol. 79, No. 6 , 1975

574

Donald W. Rogers and Nafees A. Siddiqui

As in the other dihalomethanes, CH2I2+ undergoes collisional dissociation to yield CH2I+. However the coincidence technique effectively reduces interference due to fragmentation by ion-molecule collisions. The derived daughter appearance potential is 10.55 eV. The calculated appearance potential of CHI2+ is 11.92 eV but lack of high-resolution mass analysis did not allow us to distinguish the parent ion from the H loss fragment ion. CH13.Since CHI3 has an m/e >ZOO, the ionization products were TOF mass analyzed and the coincidence scans (Figure 4) were studied. There are no literature values except for PES data2 for the ionization potential of CHI3 with which to compare our results. Since the PES data shows a broad first peak, a precise onset cannot be determined from it. The four electronic states of CHI3+ have not been identified with certainty although two are known to be degenerate. The states corresponding to the first two PES peaks are evident in our coincidence scan. From our data, the IP of CHI3 is 9.25 eV. In the threshold fit for the fragmentation process, the transition probability, p ( x ) ,was approximated by a sum of two gaussians corresponding to the second and third PES peaks. The resulting calculated onset is 9.77 eV. Acknowledgment. We are grateful for financial support from the following agencies: the U.S. Army Research Office, Durham; the Research Corporation; and the Petroleum Research Fund, administered by the American Chemical Society. References a n d Notes (1)A. W. Potts, H. J. Lempka, D. G. Streets, and W. C. Price, Phd. Trans.

Roy. Soc., Ser, A, 268, 59 (1970). (2)D. W. Turner, C. Baker, and C. R. Brundle, "Molecular Photoelectron Spectroscopy," Wiley, London, 1970. (3)A. S.Werner, B. P. Tsai, and T. Baer, J. Chem. Phys., 60,3650 (1974). (4)C. J. Noutary, J. Res. Natl. Bur. Stand., 72, 479 (1968). (5)F. P. Lossing, Bull. SOC.Chim. Be@.,81, 125 (1972). (6)T. Baer, A. S. Werner, B. P. Tsai, and S. F. Lin, J. Chem. Phys., 61, 5468 (1974). (7)T. A. Hashmall and E. Heilbronner, Angew. Chem., h t . Ed. Engl., 9,305 (1970). (8)F. Brolgi and E. Heilbronner, Helv. Chim. Acta, 54, 1423 (1971). (9)T. L. Ragle, I. A. Stenhouse, D. C. Frost, and C. A. McDowell, J. Chem. Phys., 53, 178 (1970). (IO)J. L. Franklin, J. G. Dillard. H. M. Rosenstock, J. T. Herron, K. Draxl, and F. H. Field, Nat. Stand. Ref. Data Ser., Nat. Bur. Stand., No. 26 (1969). (11) D. C. Frost and C. A. McDowell, Proc. Roy. Soc., Ser. A, 241, 194 (1957). (12)J. M. Williams and W. H. Hamill, J. Chem. Phys., 49, 4467 (1968). (13)H. Gutbier, 2.Naturforsch A, 9,348 (1954). (14)V. H. Dibeler and R. M. Reese, J. Res. Nat. Bur. Stand., 54, 127 (1955). (15) S. Tsuda, C. E. Melton, and W. H. Hamill, J. Chem. Phys., 41, 689 (1964). (16)R. I. Reed and W. Snedden, Trans. Faraday Soc., 55, 876 (1956). (17)C. A. McDowell and B. C. Cox, J. Chem. Phys., 20, 1496 (1952). (18)F. P. Lossing and G. P. Semeiuk, Can, J. Chem., 46, 955 (1970). (19)M, Krauss, J. A. Walker, and V. H. Dibeler, J. Res. Nat. Bur. Stand., Sect. A, 72, 281 (1968). (20)K. Watanabe, T. Nakayama, and J. Mottl, J. Quant. Spectrosc. Radlat. Transfer, 2, 369 (1962). (21)A. J. C. Nicholson, J. Chem. Phys., 43, 1171 (1965). (22)J. D. Morrison, H. Hurzeler, M. G. Inghram, and H. E. Stanton, J. Chem. Phys., 33, 821 (1960). (23)T. Baer, W. B. Peatman, and E. W. Schlag, Chem. Phys. Lett.. 4, 243 (1969). (24)(a) T. Baer and B. P. Tsai, J. Electron Spectrosc., 2, 25 (1973):(b) B. P. Tsai and T. Baer, J. Chem. Phys., 61,2047(1974). (25)E. G Jones, J. H. Beynon, and R. G. Cooks, J. Chem. Phys., 57, 2652 (1972). (26)T. Baer, L. Squires, and A. S. Werner, Chem. Phys., 6,325 (1974). (27)W. C. Price, J. Cbem. Phys., 4,539 (1936). (28)H. Sponer and E. Teller, Rev. Modern Phys., 13, 75 (1941). (29)V. Cermak, Collect. Czech. Chem. Commun., 33, 2739 (1968). (30) D. D. Wagman, W. H. Evans, V. B. Parker, I. Halow, S. M. Bailey, and R. H. Schumn, Nat. Bur. Stand. Tech. Note, No. 270-3 (1968). (31)R. A. Boschi and D. R. Saiahub, Mol. Phys., 24, 289 (1972).

Heats of Hydrogenation of Large Molecules. 1. Esters of Unsaturated Fatty Acids Donald W. Rogers* and Nafees A. Siddlqui Chemistry Department, The Brooklyn Center, Long lsland University, Brooklyn, New York 1120 1 (Received August 16, 1974) Publication costs assisted by the National hstitutes of Health

We have developed a hydrogen microcalorimeter for the purpose of determining heats of hydrogenation and heats of formation of large molecules, particularly those of biochemical significance. This paper reports and interprets results obtained for the seven unsaturated and polyunsaturated methyl esters of palmitoleic, palmitelaidic, oleic, elaidic, linoleic, linoelaidic, and linolenic acids. The heats of formation follow from the heats of formation of the saturated reaction products, methyl palmitate and methyl stearate, by Hess' law addition. We believe that these molecules are the largest ever studied by hydrogen Calorimetry and that the sample size on which reliable data can be obtained, about 15 hg, is among the smallest. Standard deviations of six replicate samples of each ester were about 0.3 kcal/mol per double bond.

Introduction Recently we described a hydrogen microcalorimeter1 with which we were able to determine the heats of hydrogenation of micromolar quantities of unsaturated hydrocarbons dissolved in an inert solvent. We developed this variThe Journal of Physical Chemistry, Vol. 70, N o . 6, 1975

ant on liquid-phase hydrogen calorimetry as it is usually pra~ticedz-~ for two principal reasons. The method requires very little material, making it applicable to biologically interesting substances which are commercially available in great variety but usually in small quantities and a t considerable expense. Second, the rapidity of the method offers